Reducing Processing Time of Nonlinear Analysis of Symmetric-Plan Buildings

Abstract

Nonlinear response history analysis has been a powerful tool in performance-based earthquake engineering for validating the proposed design of new structures or evaluating existing ones. When it is applied to structural systems with a large number of degrees of freedom, such as three-dimensional (3D) models of tall buildings, bridges, or dams, the analyses can be time-consuming. The prolonged computing times become more prominent in parametric studies or in incremental dynamic analyses. In order to reduce the computation time, this study proposes a practical method—a reducing time steps (RTS) procedure—whereby leading and trailing weak signals in the input acceleration record are trimmed, and the remaining record is downsampled. The test results based on several different 3D computer models of reinforced-concrete idealized structures demonstrate that the RTS method is practical, and it provides estimates of engineering demand parameters such as peak values of story drift, floor acceleration, and floor velocity within 10% of the results obtained by using the original records. The RTS procedure was further validated on three symmetric-plan steel buildings with 5, 9, and 15 stories. For all analyzed cases, the average reduction in computational time was around 50%.